🚨 Last seats available!
Multivariate Data Analysis with R & vegan led by @gsimpson.bsky.social
📅 4–7 May (onine)

www.physalia-courses.org/courses-work...

One of the things I hope to work on later this year is a package of functions for creating these structures from common R objects like sf, cubble, etc. I wrote a constructor for dataset with multiple time series the other week to see how it worked and it went ok

Excellent tweetorial by @ericjpedersen.bsky.social on Simon Wood’s new Neighborhood Cross Validation smoothness selection method, which is especially useful when fitting data with temporal and spatial dependence among observations

Prof. Aono tracked a specific species — Yamazakura (Prunus jamasakura) — in a specific place: Arashiyama, Kyoto.

We've contacted his university, local organizations, and weather services, but unfortunately none can continue the series.

We’re looking for someone who can help us continue it!

You can help if you:

🌸 Have botanical expertise and can identify Yamazakura peak bloom from photos
🌸 Are local to Arashiyama and have access to species-specific bloom reporting

Please DM me here or email me at tuna@ourworldindata.org. Even a lead helps — thank you!

I'm a data scientist @ourworldindata.org and I need help from a botanist or someone local to Kyoto, Japan! 🌸

We present one of the world’s longest climate records: 1,200 years of peak cherry blossom dates in Kyoto.

The researcher who maintained it, Prof. Yasuyuki Aono, sadly passed away last year.

I'm excited to see this out! It's also a nice coincidence, as @gsimpson.bsky.social and I just got phylogenetic smoothers working (for BM and OU processes) in our MRFtools package. I'll definitely be pointing people to your book for background.

gam-mafia.github.io/MRFtools/art...

Spooky 👻 I was just mentioning this [that @vdveenb.bsky.social had mentioned they were working on splines for gllvm] to @ericjpedersen.bsky.social a few minutes ago

Registration is now open for the Physalia online course GLLVMs: Advanced Multivariate Analysis in R (7–10 July) led by
@vdveenb.bsky.social

Join us to learn model-based ordination, JSDMs, and multispecies GL(M)Ms. Hands-on in R!

www.physalia-courses.org/courses-work...

#Ecology #RStats #GLLVM

I know there are a lot of ecologists out there who are JDSM curious, but get scared off by how technical the primary literature is...

Here's your chance to learn how they work. Strongly recommend this to anyone who is interested in species distribution modelling.

#Rstats

also why ppt presentations can become such huge files on disk.

I don't see how they could change this and still allow you resize, crop etc the original images as you edit the ppt, but it is still annoying with a image heavy ppt, & last time I used ppt the option to downscale images was not obvious

The figure wasn’t anything to do with me or my material

On the other hand, I’m kinda pissed that they couldn’t once mention me or where they got the materials from, despite citing a paper for a figure they reproduced from it.

Honestly, I’m torn; on the one hand it is absolutely a good thing that someone found these resources useful! That’s why I made them available in the first place.

That feeling when a fellow academic uses your openly licensed teaching materials for 90% of their presentation and doesn’t once comply with the CC-BY license

In an applied stats course at ugrad ( it stats major) you don’t need a lot of this stuff to fit models and evaluate them

Bottom of the Heap Live Stream YouTube video by Bottom of the Heap

I'm back up here: youtube.com/live/2lcCqjM...

Bottom of the Heap Live Stream YouTube video by Bottom of the Heap

I'm back up here: youtube.com/live/2lcCqjM...

Technical difficulties: new stream here youtube.com/live/2lcCqjM...

🎉 It's livestream day 🎉

Join me at 1700 (CET; 1600 UCT) today for two hours of GAM goodness 🤤

📽️ Youtube: youtube.com/live/A9U8e1K...
Hit the Notify me 🔔 to get a reminder when I go live

#RStats #mgcv #statistics #GAMs #DataScience 🧪

They’re mostly not myths, though it depends what one means by a loop; apply() is not slow despite using a for() internally, because it allocates storage upfront, but naive users of for() etc often ignored or don’t know to do the pre-allocation and that resulted in dog slow code 🤷🏻‍♂️

Strong 2010s Twitter #RStats vibes in this thread

Winds 2020 (2021)

#rstats and #ggplot2

I’m just so happy to hear this wonderful news Anson

Self-feedback loops matter. The model drafts, critiques itself, retrieves more context if needed, and refines. Everything is open-source: models, code, data store, and a new benchmark (ScholarQABench). 3/3 www.nature.com/articles/s41...

The citation problem is severe: GPT-4o fabricates 78-90% of citations when asked to cite recent papers. OpenScholar achieves citation accuracy on par with human experts by grounding responses in 45M open-access papers. 2/3

I've always been a fan of what the Allen Institute is doing. New in Nature: OpenScholar, an 8B RAG model for scientific literature, outperforms GPT-4o by 6% on correctness. Experts preferred its answers over human-written ones 51%-70 of the time. www.nature.com/articles/s41... 1/3 🧵

inspired by CLAUDE.md, I’ve started putting markdown files named after coworkers into work code repos so I can remind them to stop doing shit to the codebase that annoys me

for some reason they’re all mad at me now, which means ill be adding commands to JEREMY.md for an attitude adjustment

Bayesian Workflow Bayesian statistics and statistical practice have evolved over the years, driven by advancements in theory, methods, and computational tools. This book explores the intricate workflows of applied Baye...

The publisher estimates the Bayesian Workflow book will ship in June www.routledge.com/Bayesian-Wor...